During times of a natural disaster or other emergency, developers could build a ticker-like feed on their outlet’s website that streams latest notifications just from those verified accounts.
Services targeting a particular language group could do a similar search, filtering just for social media accounts available in, for example, Spanish.
2. Identifying RFP and Tender Opportunities
Devs could keep track of emerging government purchasing and procurement opportunities by searching for relevant tags amongst Government accounts and then subscribing to those social media accounts. By using text analysis APIs on those accounts to mine for terms like procurement, RFP or grant) or using dashboard integration APIs to build aggregate feeds into a single visual feed, devs could keep track of requests for purchase, tender, challenge and grant opportunities from across government departments in their domain of expertise.
Herman does note that in order for such a system to work, “if accounts share RFP opportunities a tag can be created to better track them.”
Also, such opportunities are usually published for projects greater than $25,000 USD. For smaller and micro-purchasing opportunities for dev teams wanting to start collaborating or contributing to the work of Government, GSA’s 18F has a number of initiatives of interest. In particular, their recent launch of a micro-purchasing platform so that devs can bid to contribute to open source code has been an initial success, and there are plans by 18F to create other marketplaces that have a lower barrier of entry for startups and tech suppliers.
3. Identifying New Open Data Supply or Freedom of Information Act (FoIA) Responses
Devs keeping a scan of data.gov or following their (verified) Twitter account can keep up to date with new open data sets added to the governments portal, which now includes state and local datasets alongside Federal sources. But occasionally, some departments and agencies may share data outside this system, or even share the response to Freedom of Information requests via Twitter or other social media.
For example, late last year, a FOI request was responded to by the FBI via Twitter. Disgraced CEO of Turing Pharmaceuticals, Martin Shkreli, (which under his leadership bought a generic medication often used to treat and AIDS-related illness and then raised its price by 5,500%) was allegedly involved in a variety of schemes aimed at cheating investors. He had also used his sizeable fortune to buy the limited edition (of one copy) of the Wu-Tang Clan’s “Once Upon a Time in Shaolin”. In December last year, when the FBI began investigating Shkreli, Andrew Wiseman requested that if the Wu-Tang Clan album was seized as part of the evidence trail, that its contents be uploaded as MP3’s and shared with the public.
The FBI responded via Twitter:
While this was a bit of fun, perhaps, more serious FOIA requests can be both requested via Twitter, and combining social media accounts of interest (sought via the Digital Registry’s API) and then coupled with a text mining API to alert for terms like “FoIA” could help surface obscure requests.
In the U.S., there are some requirements for agencies to share FoIA requested data publicly after they have made it available to the requester. (Some U.S. states like New York and South Carolina are saving administrative costs by putting FoIA requested data online.) If FoIA requesters could easily mention a government social media account and then share the response they received via FoIA, it could be added to an open data portal service, like, for example, the way Sqoop collects SEC and court data for others to use.
Michael Herz, the Arthur Kaplan Professor of Law at the Benjamin Cardozo School of Law in New York told ProgrammableWeb that FOIA does have a requirement for agencies to share requested data publicly after they have made it available to the requester. (An earlier version of this article said this was not the case.) However, it is a little unclear how much agencies would be held to account for not sharing FOIA respoonses publicly. Herz explains:
Agencies must post various items whether someone has asked for them or not — staff manuals, adjudicatory decisions, and so on. These items include “copies of all records, regardless of form or format, which have been released to any person [pursuant to a FOIA request] and which, because of the nature of their subject matter, the agency determines have become or are likely to become the subject of subsequent requests for substantially the same records.”
Herz says this is stipulated in U.S. Code Title 5, section 552(a)(2)(D). The following section (552(a)(2)(E) says the agency also has to prepare an index to such materials. Herz adds:
In other words, if the agency gives something to a requester, and at least two other people have also requested it or the agency thinks at least two other people will request it, then it has to include that record in the (electronic) reading room. Of course, compliance with this obligation is inconsistent, and it is not at all the same as a requirement that says “post anything provided in response to a FOIA request.”
Over time, it would be hoped that perhaps USA Government agencies could share their FoIA responses by announcing them on Twitter or other platforms and make the data available to everyone, as Herz explains is the intention of the current law governing FOIA reuqests. Being able to use an API to identify verified Government accounts and a text mining API to identify FoI mentions could help make Government more accountable, or distribute crucial but often hidden information more widely.
4. Creating a Newsreel or Streaming Service
Using the Social Registry API to identify YouTube accounts and then using YouTube APIs, a developer could create a newsreel of recent Government announcements, either for all of the US Government or for specific audiences, either by language, or by interest area.